Atmosphereic drag issues

[raidernick]

I am having an issue with a command pod I am working on. I cannot get the drag model to work properly on it. For some reason it seems unaffected by atmospheric drag in any way. For exmaple, when in the atmosphere a typical pod will "right" itself in a stable direction based on its shape. My pod does nothing. It acts the same way it does in a vacuum. It doesn't spin out of control, it just sits in whatever direction it was in before entering atmosphere and doesn't move.

Is there something I am missing besides the configuration options? Do colliders affect this, or is there something I need to do in unity?

I want the "red" side to be pulled facing prograde as that is where the heat shield attaches too. Also, the cabling and vzor jettison before entering so those won't affect drag.

mass = 1.2
dragModelType = default
maximum_drag = 0.20
minimum_drag = 0.2
angularDrag = 2

[nli2work]

without FAR/NEAR, what you have is pretty normal behavior for stock aerodynamics.

[raidernick]

Actually that is not the case. I don't use far/near and have never had this issue. Every pod from every mod I've ever downloaded (including stock ones) always "pulls" to a specific orientation during reentry.

Edited August 24, 2014 by raidernick

[Justin Kerbice]

Actually that is not the case. I don't use far/near and have never had this issue. Every pod from every mod I've ever downloaded (including stock ones) always "pulls" to a specific orientation during reentry.

That's happen when SAS is off and the pod align itself to its prograde vector. You may look in this way (turn off SAS, as you don't talk about it) or the orientation of your object (even you think the top is on top, if its not on y axis, it will not be considered as the top by KSP)

[raidernick]

I thought it was implied SAS is off, of course I wouldn't be posting here for something as trivial as that. SAS is off and the pod isn't being affected AT ALL by the atmosphere. This means you can give it a quick spin in any direction and it will continue to spin just as it would in space. It never slows down or orients itself in a specific way.

"even you think the top is on top, if its not on y axis, it will not be considered as the top by KSP", don't understand what you are trying to say here.

[nli2work]

test it yourself. stock aerodynamics, with only a pod, no additional parts, SAS off, during reentry you can orient it however you want and it will stay at that orientation regardless of drag. add some other physics significant part to it, say a parachute (but don't deploy it), then it starts to be affected by drag forces, but it has nothing to do with the pod's config settings, it's just KSP simulating two connected points experiencing drag, the parachute and pod; vs 1 point experiencing drag, just the pod. But in the end the parts are just points. orientation don't matter when the whole system is a single part.

[raidernick]

I don't need to test it myself. I've been playing for years with stock aero and have never encountered this issue. The pod orientation is always affected by drag.

[Starwaster]

I don't need to test it myself. I've been playing for years with stock aero and have never encountered this issue. The pod orientation is always affected by drag.

If it's just the pod, with NO other parts attached to it to affect drag, then the only thing that will affect orientation is center of mass.

The part's center of mass is (by default) the mesh's origin point. If you want that to change then you can either edit the mesh and re-export it or you can define CoMOffset in your part configuration file (offset is in meters, adjust as desired in the direction you want the part to orient)

Edit: Also. Important: Delete any line that says PhysicsSignificance = 1

Edit 2: Try CoMOffset = 0, -0.5, 0 to start with and adjust as needed.

Edited August 24, 2014 by Starwaster

[raidernick]

If it's just the pod, with NO other parts attached to it to affect drag, then the only thing that will affect orientation is center of mass.

The part's center of mass is (by default) the mesh's origin point. If you want that to change then you can either edit the mesh and re-export it or you can define CoMOffset in your part configuration file (offset is in meters, adjust as desired in the direction you want the part to orient)

Edit: Also. Important: Delete any line that says PhysicsSignificance = 1

Edit 2: Try CoMOffset = 0, -0.5, 0 to start with and adjust as needed.

Thank you, I thought that it had to be a COM issue as well. However, I am using bobcats COM mover plugin. When landing the pod it moves the COM down under the pod instead of the center as it usually is. I thought that this would help but it doesn't seem to make a difference. I know the COM is in fact changing as the camera moves and if you have a parachute open it affects the pods movement. I'm wondering if the game itself somehow isn't registering this change?

I know that adding a second part to the top will create drag and help pull the pod in the correct direction but I wanted to avoid adding any extra parts to the mod. With no other choice I will have to split the pod up into 2 pieces to fix this issue.

Edited August 24, 2014 by raidernick

[Starwaster]

Thank you, I thought that it had to be a COM issue as well. However, I am using bobcats COM mover plugin. When landing the pod it moves the COM down under the pod instead of the center as it usually is. I thought that this would help but it doesn't seem to make a difference. I know the COM is in fact changing as the camera moves and if you have a parachute open it affects the pods movement. I'm wondering if the game itself somehow isn't registering this change?

That's probably the case, yeah :/

[raidernick]

For now I added a quick part to the top for drag so I can continue testing other aspects of it. I will look into changing the COM in blender, I don't like using offsets in the cfg if I can help it.

[NathanKell]

raidernick: in stock "aero"dynamics, drag is purely a function of mass and therefore does not change based on part orientation. Therefore moving the center of mass does absolutely nothing, because drag does not change based on part orientation and so the center of drag will move with it.

[raidernick]

So I guess I'll have the split it up then. Not what I wanted to do but I'll just split the docking ring off the top and add high drag to that so the pod pulls into the correct orientation.

[NathanKell]

Honestly it'd be great if you split it up into lots of parts.

Main reentry module + parachute part + docking ring at top + antenna + RCS +....